Victoria’s very strange upper house

Further to Sunday’s post on the Victorian election, yesterday saw the Legislative Council, or upper house, decided (official results are here). Labor won 18 of the 40 seats, a gain of four from 2014, against ten Liberals (down four), one National (down one), one Green (down four), and ten representatives of minor parties – more (much more) about them shortly.

But first some history. The Legislative Council was designed to serve the interests of conservative property owners, in contrast to the democracy of the lower house. It was elected on a restricted franchise until 1950, and badly malapportioned until the 1984 redistribution. Apart from a few months in 1985, Labor never held a majority in the Council until Steve Bracks’s landslide victory in 2002.

The Bracks government then reformed the Council to introduce proportional representation. But it did a poor job of it.

One reason for that was understandable. It used the system of automatic preferencing by group voting tickets, which was then standard practice. This was before the 2004 Senate election demonstrated how the system could be gamed, and it produced increasingly strange results from then onwards (including 2014 in Victoria) until it was abolished for Senate elections in 2016.

The Andrews government, however, declined to follow suit, so the group voting tickets remained in effect in Victoria this year, with absurd consequences.

The second problem was more Bracks’s fault. Instead of electing Council members from the whole state as one electorate, as New South Wales and South Australia had done, Victoria was divided into eight regions, each electing only five members. That meant a quota for election of 16.7%, which severely limited proportionality.

To some extent, these two problems cancelled each other out. The relatively small regions created a big hurdle for minor parties, but group ticket preferencing allowed those parties to pool their votes, effectively setting up a lottery in which a minor party that was skilful (or lucky) with its preference deals could win seats from a very low primary vote.

Hence the infamous Glenn Druery, who, while employed as a staffer for Senator Derryn Hinch, has also practised as a preference entrepreneur, brokering deals between a diverse range of minor and micro parties. Five different parties that had apparently dealt with Druery won nine seats between them – including, perhaps not coincidentally, three for Hinch’s Justice Party, off just 3.7% of the vote. (The tenth minor party seat went to Fiona Patten’s Reason Party, which had been in Druery’s stable in 2014 but this time struck out on its own.)

Kevin Bonham and Ben Raue both have very good presentations of the figures; I won’t bother to recite them here. But they include such bizarre results as Transport Matters (a taxi industry lobby group) winning a seat in Eastern Metropolitan from 0.6% of the vote, beating the Greens who had 9.0%, and the Liberal Democrats winning a seat for their worst-performing candidate, in South-Eastern Metropolitan, off 0.8% – where the Liberals, with 29.0%, were unable to win a second seat.

But it won’t do to blame Druery for everything. If you set up perverse incentives, it’s a sure bet that someone will be found to capitalise on them. The problem is in the system, not the individual.

And despite the talk of a “lottery”, such a system is not ideologically neutral. It favors the morally unscrupulous, who are willing to detach preferencing from policy or principle, and in today’s climate those people tend to be found on the right. Hence the collective vendetta of the Druery parties against the Greens, who as a result secured only one seat for their 9.3% of the vote.

There’s nothing wrong with the idea that, say, the Animal Justice Party, with 2.5% of the vote, should have representation in the upper house, or even that Hinch’s lot should win two seats (the third is clearly over the top). But it shouldn’t depend on these completely arbitrary preference flows, insulated from voter knowledge and influence.

And why should a micro-party like Sustainable Australia (0.8%) win a seat while the DLP (2.1%) misses out? No sane person thinks that has anything to do with the actual preferences of the voters.

Proportional representation isn’t hard; lots of countries do it with very little fuss. The New South Wales and South Australian upper houses do much better than this. Calculating a Gallagher index, the standard measure of electoral unfairness, I got about 8.7 for the Legislative Council. That’s not dreadful – the lower house is about 16 (higher numbers mean more unfair, to a theoretical maximum of 100) – but it compares badly with, say, last year’s election in New Zealand at 3.5, or the last Norwegian election, with 2.7.

Since we already have a house with single-member districts that determines who governs, there’s simply no excuse for selling proportionality short in this way. Let’s get rid of the regions (or at least make them much larger), and get rid of the crazy system of group voting tickets.

I’ll leave Bonham with the last word – not only is he a very fine analyst of how systems work, but he cares about democracy:

Australia has a proud record of getting elections right. We are the home of preferential voting, the first federal election under which occurred 100 years ago tomorrow. In a world where some supposedly great but in fact badly flawed democracies are still in the nappies of first-past-the-post, we provide a shining light to places like Maine, which recently led the US by adopting preferential voting. We should take our electoral leadership role seriously and get rid of the farce that is group ticket voting.